From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dong Liu Subject: Re: cpu stall and hyperthread Date: Tue, 03 Jul 2012 16:10:46 -0400 Message-ID: <4FF351C6.3000100@gmail.com> References: <4FEBCDDE.60503@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit To: linux-rt-users@vger.kernel.org Return-path: Received: from mail-ey0-f174.google.com ([209.85.215.174]:44764 "EHLO mail-ey0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754567Ab2GCUKk (ORCPT ); Tue, 3 Jul 2012 16:10:40 -0400 Received: by eaak11 with SMTP id k11so2655555eaa.19 for ; Tue, 03 Jul 2012 13:10:39 -0700 (PDT) In-Reply-To: <4FEBCDDE.60503@gmail.com> Sender: linux-rt-users-owner@vger.kernel.org List-ID: Hi, After reading the RCU cpu stall doc. I thought that CONFIG_NO_HZ might have something to do with this problem. And I think the NO_HZ config is not good for RT performance, so I made a new kernel without NO_HZ. It lasted longer but eventually cpu stalls happened again. I then disable the cpu stall checking by set /sys/module/rcutree/parameters/rcu_cpu_stall_suppress to "1", but it did not solve the problem. Instead of cpu stall I had BUG: soft lockup - CPU#0 stuck for 279s! [swapper/0:0] And even I disabled Hyper-Threading, I also got cpu stall now. CPU stalls also occurred in KVM guest OS as well. But I never get cpu stall when I'm running the normal kernel without -rt patch. Is the cpu stall a known issue with the rt patch? Thanks, Dong On 6/27/12 11:22 PM, Dong Liu wrote: > Hi, > > I'm running 3.4.3-rt11 and 3.2.18-rt29 on a system with Intel I7 860, I > noticed that with hyperthread enabled, I'm getting kernel error > > rcu_preempt detected stalls on CPUs/tasks: { 3} (detected by 5, t=60002 > jiffies) > > But with hyperthread disabled, no such error. > > Is preempt_rt not supposed to be used with hyperthread? > > Thanks, > > Dong > > P.S.